Attempts to develop effective preventive health programs are severly limited by one serious drawback: We know very little about the origins of health beliefs and health behaviors. This proposal is aimed at correcting that deficiency by studying the origins of the health beliefs and behaviors of children, adolescents, and young adults. Studies 1 and 2 are secondary analyses of large nationally representative samples of (respectively) 7-11 and 12-17 year-old children, their parents and teachers. These two studies will focus onparental influence on children's health beliefs and behaviors. Study 3 is a 2 1/2 year longitudinal stuldy of 18-21 year olds. Respondents will be contacted six times at approximately half year intervals during the course of the study. Information from parents and peers will also be gathered. In addition, two experiments on media effects will be conducted during the latter two years of the study. Structural equations analyses will be used in all three studies to model the process whereby people develop health beliefs and health behaviors. Specific hypotheses will test (1) the relative importance of parents, peers, the media, and direct experience with disease in determining health beliefs and behaviors; (2) the relative importance of (a) general and specific health beliefs, and (b) specific health beliefs based on prior behavior, in determining health behaviors; and (3) the generality, stability, origins, and consequences of common illness schemas. Throughout the proposal a general social psychological perspective is taken. The ultimate long-run objective of the proposal is to aid the development of cogent, theoretically based preventive health programs.